1 68 



Transactions of the Canadian Institute. 



[vol. IX 



The polonium which was carried by the electrode B could be made to 

 face in any direction by turning the axial rod A. This rod passed through 

 and was sealed to a glass tube D which fitted over another glass tube E 

 so arranged that the joint could be covered with mercury and so made 

 air tight. The electrode C which was made of carbon, was connected 

 with the quadrant of an electrometer, the connecting rod passing through 

 an ebonite plug fitted into the side of a vessel by a side tube F. Another 

 side tube G connected the vessel to the McLeod Gauge and the Gaede pump. 

 The vessel itself, which was a cylindrical brass tube about 5 cm. in diam- 

 eter and 8 cm. in length was connected to earth. All joints were made 

 air-tight by means of wax and solder. 



The first experiment was conducted as follows. A fresh carbon 

 electrode was placed in the apparatus, the polonium was turned away from 

 the electrode and the air was pumped from the vessel for over an hour 

 with the pump going continuously. Then the polonium was charged to 

 a positive potential of 80 volts and turned so that it faced the carbon 

 electrode. A reading was at once taken of the rate at which the carbon 

 electrode charged up, and similar readings were made at different in- 

 tervals of time afterwards. The results obtained are given in Table XII. 



Table XII. 



Fresh carbon in apparatus. 

 Turned polonium to face carbon i hour, 20 minutes after pump was started. 

 Voltage on polonium — 80 volts. 



These results shew that there was no sign of a "fatigue" in the rate 

 at which the carbon electrode charged up with the time, for the rate 

 was practically constant throughout all the readings. 



We have seen therefore that it is possible to get rid of the fatigue 

 effect altogether by withdrawing the gas from the surface of the carbon 

 before beginning the bombardment by « rays. The "fatigue" then 

 which has been described in the previous sections must be due to a de- 

 crease in the secondary radiation from the gas occluded at the surface 

 of the brass or carbon electrode used. 



